AHA: Diuretic-based therapy superior to ultrafiltration in HF patients

Stepped administration of pharmocologic therapy was superior to ultrafiltration in safely achieving weight loss and change in serum creatinine levels in patients experiencing decompensated heart failure (HF) with cardiorenal syndrome, according to CARRESS-HR results presented Nov. 6 at the American Heart Association's (AHA) scientific sessions in Los Angeles, and simultaneously published online in the New England Journal of Medicine.

Patients with acute decompensated HF often are treated with intravenous diuretics to achieve proper fluid balance. However, up to one third of patients with decompensated HF experience cardiorenal syndrome, a worsening of kidney function. Administration of diuretics may contribute to kidney injury. Venovenous ultrafiltration is sometimes used as an alternative to diuretics for patients with compromised renal function, as well as patients who do not respond to pharmacological management.  

Bradley A. Bart, MD, of the Hennepin County Medical Center in Minneapolis, and colleagues designed the CARRESS-HF (Cardiorenal Rescue Study in Acute Decompensated Heart Failure) to compare the safety and efficacy of ultrafiltration and stepped pharmacologic therapy for patients with HF, compromised renal function and persistent congestion.

The study enrolled 188 patients hospitalized with a primary diagnosis of acute decompensated HF between June 22, 2008, and Jan. 27, 2012, at 22 sites in the U.S. and Canada. To be eligible for enrollment, all patients experienced at least two of the following conditions: at least two peripheral edema, jugular venous pressure greater than 10 cm of water or pulmonary edema or pleural effusion on chest radiography. Patients were excluded if they had a serum creatinine level exceeding 3.5mg/dL at time of admission or if they were receiving intravenous vasodilators or inotropic agents.

The median age of the enrolled patients was 68; 75 percent were male; 85 percent had high blood pressure; 66 percent had diabetes mellitus; and the mean ejection fraction was 33 percent.  

The patients were randomized to receive either stepped pharmacological therapy or ultrafiltration. The patients underwent ultrafiltration using Aquadex System 100 (CHF Solutions), which removed fluids at a rate of 200 ml per hour. Patients in the pharmacologic group received intravenous administration of a diuretic to manage signs and symptoms of congestion, at doses necessary to maintain urine output of three to five liters a day. Treatment was continued in both groups until signs and symptoms of congestion were reduced to the extent possible. The study endpoints were change in serum creatinine level and change in weight over the first 96 hours after randomization.

In the pharmacologic therapy group, the mean duration of treatment was 92 hours. Therapy was discontinued because the best possible fluid volume was reached in 72 percent of the patients. Other reasons for the discontinuation of therapy were: increase in creatinine level (12 percent); intravascular volume depletion (3 percent); and blood pressure decrease/clinical instability (2 percent).

The mean duration of ultrafiltration was 40 hours. Ultrafiltration achieved the best possible fluid volume in 50 percent of patients. Treatment ended for other patients because: creatinine levels rose (16 percent), vascular access problems developed (9 percent), and the patients experienced ultrafiltration circuit thrombosis (9 percent).

At 60-day follow-up, patients receiving ultrafiltration had higher rates of serious adverse events than the pharmacologic group (72 percent vs. 57 percent). The ultrafiltration group experienced higher rates of kidney failure, bleeding and catheter-related complications. Mortality was 17 percent in the ultrafiltration group and 13 percent in the pharmacologic group. Composite rates of death and rehospitalization for heart failure were 38 percent in the ultrafiltration group and 35 percent in the pharmacologic group; composite rates of death and rehospitalization for any cause were 61 percent in the ultrafiltration group and 48 percent in the pharmacologic group.

The study was designed to enroll 200 patients but enrollment ended early on the recommendation of the independent data and safety monitoring board due to the lack of evidence of benefit and rate of adverse events in the ultrafiltration group.

Both groups experienced similar weight loss, but patients in the ultrafiltration group experienced significant increases in creatinine levels. Creatinine levels in the pharmacologic group went down. The authors explained that the duration of ultrafiltration was longer in this trial than in other previous trials, which may have contributed to the increase in creatinine levels at 96 hours. They pointed out that at 30 days and 60 days, the mean creatinine levels were below baseline for both groups.

In an accompanying editorial, W.H. Wilson Tang, MD, at the Cleveland Clinic, noted that the decrease in serum creatinine levels among the pharmacologic group was "unexpected," but it gave credence to the idea that worsening kidney function among decompensated HF patients may be due to "diminished renal reserve, rather than treatment effects."

Wilson suggested that the rate at which congestion is alleviated may have an impact on creatinine levels, and that slower but steady ultrafiltration may relieve congestion more safely.

Noting that ultrafiltration was once considered a promising, potentially beneficial alternative for cardiorenal syndrome rescue, it "can no longer be considered to be a favorable choice for routine therapy" for these patients. He asserted that more research is necessary to prevent the development of cardiorenal syndrome is necessary, and raised the possibility that aggressive diuresis may be a contributing factor.